TYK2-STAT1-BCL2 pathway dependence in T-cell acute lymphoblastic leukemia

Abstract

Targeted molecular therapy has yielded remarkable outcomes in certain cancers, but specific therapeutic targets remain elusive for many others. As a result of two independent RNA interference (RNAi) screens, we identified pathway dependence on a member of the Janus-activated kinase (JAK) tyrosine kinase family, TYK2, and its downstream effector STAT1, in T-cell acute lymphoblastic leukemia (T-ALL). Gene knockdown experiments consistently showed TYK2 dependence in both T-ALL primary specimens and cell lines, and a small-molecule inhibitor of JAK activity induced T-ALL cell death. Activation of this TYK2-STAT1 pathway in T-ALL cell lines occurs by gain-of-function TYK2 mutations or activation of interleukin (IL)-10 receptor signaling, and this pathway mediates T-ALL cell survival through upregulation of the antiapoptotic protein BCL2. These findings indicate that in many T-ALL cases, the leukemic cells are dependent upon the TYK2-STAT1-BCL2 pathway for continued survival, supporting the development of molecular therapies targeting TYK2 and other components of this pathway.

Significance: In recent years, "pathway dependence" has been revealed in specific types of human cancer, which can be important because they pinpoint proteins that are particularly vulnerable to antitumor-targeted inhibition (so-called Achilles’ heel proteins). Here, we use RNAi technology to identify a novel oncogenic pathway that involves aberrant activation of the TYK2 tyrosine kinase and its downstream substrate, STAT1, which ultimately promotes T-ALL cell survival through the upregulation of BCL2 expression

Figure 1. TYK2 is required for the survival of T-ALL cells

A, Primary cells from a pediatric patient with T-ALL were electroporated with siRNAs targeting each member of the tyrosine kinome as well as N-RAS, K-RAS, and control scrambled siRNA. Cell viability values are shown in relation to the group mean ± two standard deviations (SD). Values above or below the 2-SD threshold were considered significant. B, shRNA screen with approximately 5,000 inducible shRNAs that collectively target 1,740 genes were performed on three T-ALL cell lines (JURKAT, CCRF-CEM and SKW-3/KE-37) and four diffuse large B-cell lymphoma (DLBCL) cell lines (Ly3, Ly10, Ly7 and Ly19). Depletion of TYK2 shRNA from the cell population was calculated as shRNA-uninduced/induced (log2), and is shown as the mean ± standard error of the mean (s.e.m.) of four independent experiments. C, Validated shRNAs targeting JAK1, JAK2, JAK3 or TYK2 as well as two control shRNAs (GFP and Luc) were transduced by lentivirus infection into JURKAT cells. The number of shRNAs tested is indicated in parentheses. Cell viability was measured after 3 and 7 days of infection. Growth rate (day 7/day 3) relative to the mean of control samples is reported as the mean ± s.e.m (n=2-6). * P<0.05 by two-sample, two-tailed t-test. D, The three TYK2-targeting shRNAs as well as control GFP shRNA were transduced by lentivirus infection into JURKAT cells. Relative cell growth values (means ± s.e.m of triplicate experiments) at days 3, 5, 7, and 9 after infection are shown. E, The three TYK2-targeting shRNAs as well as control GFP shRNA were transduced in five T-ALL cell lines (JURKAT, RPMI-8402, HPB-ALL, MOLT-4 and LOUCY). Growth rate (day 7/day 3) relative to control is shown as the mean ± s.e.m of triplicate experiments. * P<0.05, ** P<0.01, *** P<0.001 by two-sample, two-tailed t-test. F, Primary T-ALL cells were initially expanded by primagraft into Rag2^−/−γc^−/− or NOD/Scid/Il2rg^−/− mice, subsequently expanded on OP9-DL1 or MS5-DL1 cells, and electroporated with non-specific siRNA (control) or siRNA targeting TYK2 followed by a 4-day culture. Values represent mean percent cell viability (normalized to viability of control siRNA) ± s.e.m of quadruplicate experiments. * P<0.05 by two-sample, two-tailed t-test. G, JURKAT, RPMI-8402, HPB-ALL or LOUCY cells harboring GFP or TYK2 shRNAs were analyzed for rate of apoptosis after 4 days of lentiviral infection by flow cytometric analysis of cells stained with Annexin V-FITC. The values are means ± s.e.m of triplicate experiments. * P<0.05, ** P<0.01 by two-sample, two-tailed t-test. H, cDNA containing the wild-type (WT) TYK2 was transduced by retroviral infection in JURKAT cells. The cells were then transduced by lentivirus infection with control GFP or TYK2 #2 shRNA, which targets the 3′UTR of TYK2 mRNA. Whole cell extracts were harvested and subjected to immunoblot analysis with antibodies specific for total TYK2, PARP and α-tubulin (internal control). Growth rate (day 7/day 3) relative to control is shown as the mean ± s.e.m of triplicate experiments. *** P<0.001 by two-sample, two-tailed t-test.

Figure 2. TYK2-STAT1 pathway upregulates the anti-apoptotic protein BCL2 in T-ALL

A, Diagram of proposed TYK2 pathway. B, JURKAT cells expressing an empty vector or WT TYK2 cDNA were transduced with TYK2 or control GFP shRNA. Whole cell extracts were harvested and subjected to immunoblot analysis with antibodies specific for total TYK2, phospho-STAT1 (Y701), STAT1, phospho-STAT3 (Y705), STAT3 and α-tubulin. C, The two STAT1-targeting shRNAs as well as control GFP shRNA were transduced into JURKAT, HPB-ALL and LOUCY cells. Cell viability was measured after 3 and 7 days of infection. Growth rate (day 7/day 3) relative to control is shown as the mean ± s.e.m of triplicate experiments. * P<0.05, *** P<0.001 by two-sample, two-tailed t-test. D, Global gene expressions in JURKAT cells transduced with TYK2, STAT1 or control shRNAs (GFP and Luc) were measured by microarray analysis. The genes significantly downregulated by TYK2 or STAT1 knockdown (KD) compared to control were determined and used as gene sets for the gene set enrichment analysis (GSEA). GSEA plots indicate the degree to which genes are overrepresented at the extreme left (downregulated by KD) or right (upregulated by KD) of the entire ranked list. Solid bars represent genes. Normalized enrichment score (NES) and P-values are indicated. E, BCL2 mRNA expression levels in each shRNA-transduced samples were determined by microarray analysis. Values are means ± s.e.m. of duplicate experiments. *** P<0.001 by two-sample, two-tailed t-test. F, The TYK2 shRNA as well as control GFP shRNA were transduced into T-ALL cell lines. BCL2 mRNA expression was measured by quantitative RT-PCR and normalized by GAPDH expression. Gene expression changes (KD/control) were shown as mean ± s.e.m. of duplicate experiments. G, The WT or kinase-dead form of TYK2 cDNA was transduced into MOLT-4 cells by retrovirus infection. The selected clone was then transduced with GFP or TYK2 shRNA by lentivirus infection. Whole cell extracts were subjected to immunoblot analysis with antibodies specific for TYK2, BCL2 and α-tubulin. H, Apoptosis was measured after 4 days of infection by flow cytometric analysis of cells stained with AnnexinV-PE. Values are means ± s.e.m. of duplicate experiments. *** P<0.001 by two-sample, two-tailed t-test. I, The WT or Y701F STAT1α cDNA was transduced in JURKAT cells by retrovirus infection. The selected clone was transduced with GFP or STAT1 shRNA by lentivirus infection. Whole cell extracts were subjected to immunoblot analysis with antibodies specific for total STAT1, phospho-STAT1 (Y701), BCL2 and α-tubulin. Arrowhead indicates STAT1α isoform. J, Apoptosis was measured after 4 days of infection by flow cytometric analysis of cells stained with AnnexinV-PE. Values are means ± s.e.m. of duplicate experiments. * P<0.05 by two-sample, two-tailed t-test.

Figure 3. Activating mutations of TYK2 gene in T-ALL

A, Diagram of TYK2 functional domains with locations of point mutations (arrows) that were identified in T-ALL specimens. Amino acid (a.a.) positions are indicated. B, Crystal structure of TYK2 depicting the positions of two of the tumor-associated, activating mutations (E957D and R1027H) within the TYK2 kinase domain (red spheres). Also shown is the location of the M978 residue that impairs TYK2 kinase activity when mutated to tyrosine or phenylalanine (blue sphere). C, Ba/F3 cells infected with retrovirus expressing mutant or WT TYK2 cDNA were cultured in the absence of IL-3 for 16 days, with cell density measurements made daily. Total cell numbers, plotted as means ± s.e.m of triplicate experiments, are shown. D, Ba/F3 cells expressing WT TYK2, different TYK2 mutants or the positive control mutant TYK2-V678F, all in the absence of IL-3 or WT TYK2 in the presence of IL-3, were subjected to immunoblot analysis with antibodies specific for total or phospho-TYK2, STAT1, STAT3, STAT5 and ERK1/2 as well as β-actin (internal control). E, Ba/F3 cells expressing TYK2-E957D, TYK2-E957D/M978Y, or TYK2-E957D/M978F cDNA were cultured in the absence of IL-3 with cell density measurements made daily. Total cell numbers were plotted as the means ± s.e.m of triplicate experiments. F, Single and double TYK2 mutants were transiently transfected into HEK293 cells, and whole cell extracts were subjected to immunoblot analysis with antibodies specific for total and phospho-TYK2, STAT1, STAT3 and STAT5 as well as β-actin. G, Murine bone marrow cells were infected with retrovirus expressing empty vector, TYK2-WT, or TYK2-E957D cDNA. The infected cells were co-cultured with OP9 DL1 cells in the presence of IL-7 and FLT3 ligand. Viable cells from these cultures were counted by propidium iodide exclusion, as were cells from wells containing only OP9 DL1 stromal cells (to establish a gate that excludes OP9 DL1 cells from the counts). Values are the mean fold-changes ± s.e.m. relative to day 2 (n=3).

Figure 4. IL-10 receptor signaling is required for TYK2 activation and cell survival in T-ALL

A, Validated shRNAs targeting IL10RA, IL10RB or IL10 as well as control GFP shRNA were transduced by lentivirus infection into four T-ALL cell lines (HPB-ALL, JURKAT, MOLT-4 and LOUCY). Relative cell growth at days 3, 5, 7 and 9 after infection was evaluated. Values are means ± s.e.m. of triplicate experiments. B, Apoptosis was measured after 6 days of infection by flow cytometric analysis of cells stained with AnnexinV-FITC. Values are means ± s.e.m. of duplicate experiments. ** P<0.01, *** P<0.001 by two-sample, two-tailed t-test. C, BCL2 mRNA expression level was measured by quantitative RT-PCR and normalized by GAPDH expression. BCL2 expression relative to control is shown as the mean ± s.e.m. of duplicate experiments. D, Whole cell extracts harvested from HPB-ALL or JURKAT cells transduced with shRNAs targeting control GFP, IL10RA or IL10RB were subjected to immunoblot analysis with antibodies specific for total STAT1, phospho-STAT1 (Y701) and α-tubulin.

Figure 5. Activity of small-molecule inhibitors of TYK2 against transformed Ba/F3 and T-ALL cells

A, Ba/F3 cells transformed by TEL-ABL, TEL-JAK1, TEL-JAK2, TEL-JAK3, or TYK2-E957D and cultured with graded concentrations of JAK Inhibitor I for 3 days. Cell viability values are means ± s.e.m. percentage of the untreated control values in triplicate experiments. B, LOUCY, JURKAT, MOLT-4, HPB-ALL and RPMI-8402 cells cultured with graded concentrations of JAK Inhibitor I for 3 days. Cell viability values are means ± s.e.m. percentages of the untreated control in six experiments. C, JURKAT and LOUCY cells were cultured in the absence or presence of JAK Inhibitor I (3 μM) for 7 days. The cell number was shown as fold change from day 0 for each cell line (means ± s.e.m. of duplicate experiments). D, Ba/F3 cells expressing TYK2-E957D cultured with graded concentrations of JAK Inhibitor I and subjected to immunoblot analysis with antibodies specific for phospho-TYK2, phospho-STAT1, PARP and α-tubulin. E, JURKAT, HPB-ALL, and LOUCY cells cultured with graded concentrations of JAK Inhibitor I for 24 hours, and assessed for apoptosis by flow cytometric analysis after staining with AnnexinV-FITC. Values are means ± s.e.m. percentages for duplicate experiments. * P<0.05, *** P<0.001 by two-sample, two-tailed t-test.